Calibrating the mean reverting jump diffusion model to Australian spot electricity prices

Alcock, Jamie and Goard, Joanne and Vassallo, Tony (2007) Calibrating the mean reverting jump diffusion model to Australian spot electricity prices. Australian and New Zealand Mathematics in Industry Study Group > 24th MISG [Wollongong 5/2/2007 - 9/2/2007].

Full text available as:

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

Problem Statement

Integral Energy is exploring alternative methods for modelling spot electricity prices. Spot price models are used to generate forecasts of the pool electricity price, and these forecasts are subsequently used to calculate the Earnings at Risk (EaR) for Integral Energy’s electricity portfolio.
The most frequently referenced model for this task in the current literature is the Mean Reverting Jump Diffusion (MRJD) model. While the analytic fundamentals and features of this model are well established, the issue of how accurately the model can be calibrated to historical data remains unaddressed. This issue needs to be resolved, in order to determine the suitability of the MRJD model for the purpose of generating spot electricity price forecasts.
The aim of this project is to determine a means of calibrating the MRJD model to historical Australian spot electricity prices, such that the model will accurately capture all the features of the data, and all necessary model parameters can be determined efficiently.

Archive Staff Only: edit this record